472 Agnes Arber.— The Phyllode Theory of 
and takes the view that the band-like submerged leaves of this plant are not 
reduced leaves, in which the lamina has disappeared and the petiole alone 
survives, but rudimentary leaves, in which no differentiation of blade from 
petiole has occurred, and which are 4 Hemmungsbildungen ’ (arrested stages) 
of the mature form of leaf. He supports this view by recalling that, in the 
ontogeny of the individual arrow-head leaf, stages are passed through corre¬ 
sponding, first, to the band-shaped submerged leaf and, secondly, to the 
oval floating leaf. It is true that these developmental facts are not easy to 
reconcile precisely with the phyllode theory as enunciated by de Candolle, 
but they fall readily into place when considered in the light of Henslow’s 
extension of de Candolle’s view. If the blade of Sagittaria be merely an 
expansion and development of the apical region of the petiole, the band¬ 
shaped leaf is indeed, as Goebel says, comparable with a complete air-leaf 
and not merely with its petiole. Where Henslow would part company 
with Goebel would be in regarding both the simple band-leaf and the highly 
differentiated air-leaf as homologous with the petiole alone of a typical 
Dicotyledon. 
Henslow’s corollary to de Candolle’s theory is obviously even more 
difficult to prove or disprove than the main theory itself. As we shall see in 
a later section of this paper (p. 488), it gains a considerable degree of support 
from anatomical evidence. 
5. An extension of the 4 Phyllode Theory ’ to the leaves 
of certain Gymnosperms. 
Though the Gymnosperms fall beyond the bounds of our present 
subject, it may be worth while to touch briefly upon the possibility of an 
application of the phyllode theory to their case. 1 It seems to the present 
writer an indication of the validity of this theory, that it affords a point of 
view which reveals fresh interpretations of leaf morphology in other groups 
of Seed Plants. 
Long ago Asa Gray 2 suggested, as a possible alternative to other 
explanations, that the leaves of Pines, Cypresses, &c., might be interpreted 
as homologous with petioles. He, apparently, did not pursue the matter, 
but his idea seems to the present writer a fertile one. There are many 
cases among the Gymnosperms to which such an explanation may well be 
applicable. The uniformity of leaf-structure in the Coniferae is very remark¬ 
able. Compton 3 has recently drawn attention to the close relation to one 
another of the predominant types within the group, and to the association 
of the narrow acicular, linear-lanceolate, or cupressoid leaf-forms, with the 
absence of lateral pinnation of the foliar vascular system. He remarks that 
4 emphasis should be laid on the small power of the Conifers to vary the 
1 The writer hopes to deal more fully with this subject in a later paper. 
2 Gray, A. (1887). 3 Compton, R. H. (1911). 
